Sébastien Bubeck – Talks

2013

Whenever possible I am now doing my talks on the blackboard (some of them are recorded, check videolectures.net and youtube for example).

2012

  • Optimal discovery with probabilistic expert advice. [slides] (Presented at CDC 2012.)

  • Regret analysis in finitely-armed and continuously-armed bandits. [slides] (Presented at the Department of Management Science and Information Systems, Rutgers.)

  • Two basic problems in finite stochastic optimization. [slides] [Video] (Presented at the ICML 2012 workshop Exploration and Exploitation 3; at the Department of Electrical and Computer Engineering, University of Wisconsin-Madison; at the Department of Computer Science, Rutgers; and at the Department of Electrical Engineering, Princeton University.)

  • Towards Minimax Policies for Online Linear Optimization with Bandit Feedback. [slides] [Video] (Presented at COLT 2012.)

2011

  • Tutorial on Bandits Games. [slides] (Presented at ALT 2011 and ACML 2012.)

  • Minimax Policies for Combinatorial Prediction Games [slides] [Video] (Presented at COLT 2011.)

2010

2009

  • Pure Exploration in Multi-Armed Bandit Problems. [slides] (Presented at ALT 2009.)

  • Minimax Policies for Adversarial and Stochastic Bandits.[slides] (Presented at COLT 2009.)